5-GPU Orchestration System

AI Bank

Complete AI orchestration system with multi-GPU resource distribution, automated trading via MDPS Engine, and intelligent service management.

View Architecture Quick Start Guide

GPU Architecture

Optimized resource distribution across 5 GPUs for maximum performance

GPU #0

Cloud Agents

  • • Claude (Anthropic)
  • • ChatGPT (OpenAI)
  • • Gemini (Google)
GPU #1

LM Studio

  • • Lightweight Inference
  • • Fast Response
  • • Custom Models
GPU #2

Ollama Primary

  • • DeepSeek-R1
  • • Qwen
  • • Mistral
GPU #3

Ollama Secondary

  • • Distributed Inference
  • • Load Balancing
  • • Scalability
GPU #4

OmniCore + MDPS

  • • Orchestration
  • • Trading Engine
  • • Decision Making

Service Status

API Endpoints

Connect to services via REST endpoints

Ollama

Open →
http://localhost:11434

LM Studio

Open →
http://localhost:1234

AnythingLLM

Open →
http://localhost:3001

OmniCore

Open →
http://localhost:9000

Trading API

Open →
http://localhost:9002

API Gateway

Open →
http://localhost:9100

Management Commands

Quick access to system utilities

Monitoring

aibank-monitor

Real-time service health

Gateway

aibank-gateway

Test all API endpoints

Trading

aibank-trading

Trading system status

Analytics

aibank-analytics

Performance analytics